TEA for Windows 1.0
At the autumn of 2000 I was a Windows user. It was a good time. I played Fallout 2, posted tons of messages at the Fidonet (it was still popular in ex-USSR probably until Y2K) and started to write a lot of articles as a freelance journalist. I tried some text editors, tried MS Word, but all they did not inspired me. So I decided to create one for myself.
I cannot say that I am the powerful programmer - I am not the programmer at all. I can't make 3D engines like John Carmack and I can't write the compiler or the video-codec. But the text editor, if I use the ready-to-use editor's engine - it is a relatively simple thing.
I had some Delphi programming experience, so I started my text editor with Delphi. For the first time I used only standart components. The name of the editor was "Typewriter". I used it for my writings and I was happy. Sometime soon I uploaded Typewriter to my homesite and got some feedback from users.
Typewriter had a typical MDI-interface with child windows that floated inside of the main one. Typewriter featured some text-processing functions and sounds similar to the true typewriter due to embedded stereo samples which I synthesized for myself. Also I put to the editor the brain waves synchronizer (something like Autozen program for Linux), plugins support, a simple file manager, a bookmarks engine, an ugly installer and some other stuff. Typewriter 3.0 was the prototype of TEA. Typewriter had a freeware status (but not with an open source).
At the fall of 2001 I have decided to promote my editor to the world. So I renamed Typewriter into the TEA and released the first public version - 1.0. Actually, in the first time I pronounced the new name of the editor when I went with my girlfriend to the subway entry. I said that "TEA" is sounds good and looks good too. It is short and habitual. She agreed.
TEA for Windows 1.0
I released TEA 1.0 and registered it on some Ukrainian and Russian Tucows-like servers. TEA becomed popular. I continued to add new features. I used the new text-editor engine, added tabbed layout, customizable toolbars, the file manager called Neptuno, the HTML 3.2 browser Retro, a hundred of functions. TEA had three editor engines - the plain text, the hex and the RTF. TEA was highly deeped into the object-oriented programming. I made the Virtual editor inside of TEA. That helped me to change editing engines on the fly. Among my own code, I used many third-party freeware or royalty-free libraries and components.
TEA for Windows had some disadvantages - the Russian-only interface and documentation (nearly 200 kbytes of the text!) and non-Free (but Freeware license). I can notice that TEA for Windows used the NSIS-made installer, with the 1.3 Mbytes size. It is for the "all-in-one" text processor which compared to the commercial one HomeSite.
At the year 2002 Lemon, the designer, joined to the project. He made nice artwork and TEA home site, and TEA documentation design also. Some time later, due to the disagreement between us about the accessibility of the design, he went out of the project, and I continued to make all artwork and design for myself again (and I removed all Lemon's button glyphs and icons).
2003. I used Linux more and more often, but there was no TEA available. At the same time I saw that the life time of TEA for Windows has coming to the end. Too many features. Too bloated code. In december 2003 I, with some sorrow in my soul, released the last version of TEA for Windows. It is still available for downloading here.
TEA for Windows 11.0
At the autumn of 2003 Linux becomed my default operation system. I felt the lack of TEA and decided to port it from Windows. But TEA never had been destinated for the porting. TEA was glued to Delphi and dozens of libraries. The solution was to rewrite TEA from the zero point.
When I started TEA for Linux, I knew some basic C-programming rules. That's all. I did not know nothing about the Linux programming, GTK, autoconf/automake and all that stuff. I downloaded GTK documentation. I had Gedit and Bluefish source and watched into that, how it deals with GTK functions. Then I generated a project in Glade2 and merged it with some code that I took from Gedit and Bluefish. That was a code which related to buffers-handling, statistics, and "show line numbers" function.
I started to hack around it, then I wrote a tabbed engine and saw that I can do a really usable editor. And, as usual, I began to add the features. There was a problem with a writing of English documentation. As the Russian-speaking person, I never speaks English much. I can freely read English, but as I live in the country where people uses in their conversation Russian and Ukrainian languages, I had no practice in real English, so I felt myself not very comfortable when I wrote the manual. You may see the "evolution" of my English in TEA's ChangeLog, reading it from the early records to the modern times.
I made a site, I uploaded TEA tar.bz archive. I updated it as far I added to TEA a new stuff and fixed an old one. Many good people started to make packages, translations, and send me suggestions. Also I received critics about the two things. The first one was the question, why I increase the major version number too often. But I think that it is more logical rather than uses something like tea-2.0.1.07.2-pre-04-fix. And another one "problem" was a tar.bz2 name - tea.tar.bz without any digits. And again, I do not see any problem here. But at May 2005 I gived up with a tarball name and started to call it following the TEA version.
But a critics is rare. I think that is a good sign. TEA becomes a mature project - mostly by work of contributors, because that THEY test new versions as I never can test, and because that THEY wrote translations, patches, advices and propositions, making TEA more stable and full-featured.
Currently TEA has some disadvantages. The main one is the static, non-automatic syntax highlighting. At this time I have no inspiration to write new hl-engine. Possibly in the future - yes. But for now - no. Yes I know about the GtkSourceView. But TEA will use it only when GtkSourceView becomes a part of GTK. Understand me right - one of the TEA core features is a GTK-only (and ASpell as an option) dependencies. There are many possibilities to use much of fine libraries - GtkSourceView, Enca, GnomeVFS, but if I'll do that, TEA will be too bloated, too complex for a simple building.
The future of TEA. As far I live and write texts, I shall develop TEA. Maybe I'll port it to Qt or even Mono (and don't ask what I smoke - I smoke nothing). But such thing really can happen if I'll feel the limitations of development with a pure GTK and C. Currently I totally happy with that holy pair (C/GTK).
TEA needs more translations. The Chinese one - how I dream about the Chinese translation! Also I physically can't test and fix TEA on AMD 64 CPU, but I know that Kwas is crashes when runs on it.
That's all. Drink tea, not alcohol.